Diagnosis of primary cutaneous B-cell lymphoma (PCBCL) is supported by the demonstration of a monoclonal B-cell population. Immunoglobulin heavy chain (IgH) gene-rearrangement analysis by polymerase chain reaction (PCR) is a reliable technique to detect B-cell monoclonality in paraffin-embedded tissue, but the presence of numerous reactive B lymphocytes in PCBCL may complicate the interpretation of clonality test results. To test this hypothesis, IgH gene-rearrangement analysis by PCR was performed on paraffin-embedded whole tissue sections of 19 cutaneous B-cell infiltrates diagnosed either as consistent with PCBCL (10 specimens) or unclassified lymphoid infiltrates (ULI) (9 specimens). In specimens that did not show monoclonal bands by IgH gene-rearrangement on DNA extracted from whole tissue sections, clonality assays were repeated on microdissected B-cell subpopulations suspicious for neoplastic cells. In the analysis of whole tissue sections, 4 (40%) of 10 specimens consistent with PCBCL showed one or two monoclonal bands, whereas 9 of 9 ULI specimens showed either a ladder or a smear. Clonality analysis of microdissected B-cell subpopulations showed 3 additional PCBCL specimens (total, 7 of 10) and 1 ULI specimen (total, 1 of 9) with unequivocal and reproducible monoclonal bands. Addition of microdissection increases the sensitivity of PCR-based B-cell clonality assay in PCBCL compared with analysis performed on the whole section (70% versus 40% monoclonal cases) and allows the recognition of a dominant clone in ULI specimens, possibly representing early PCBCL.